Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster E in Poster Session E: Thursday, August 6, 10:30 am – 12:15 pm, Kimmel Center, Shorin & Rosenthal Rooms

The hippocampus represents abstract graph structure at multiple scales along its longitudinal axis

Valerio Rubino¹, Anna-Lea Beyer², Charley M Wu², Manuela Piazza¹; ¹University of Trento, ²Technische Universität Darmstadt

Presenter: Valerio Rubino

Humans extract latent structure from their environment, and growing evidence suggests the hippocampus encodes such structure through predictive representations. In spatial domains, these representations are organized along the hippocampal longitudinal axis, with posterior regions encoding finer-grained, local structure and anterior regions supporting more global, coarse-grained structure. To test whether the same principle applies to abstract, non-spatial relational structures, we reanalysed fMRI data from participants viewing sequences drawn from a previously learned graph. We found that the left hippocampus represents abstract graph structure at multiple scales, with predictive scale increasing from posterior to anterior regions. Our findings suggest that hippocampal multi-scale predictive representations may be a domain-general mechanism for relational processing.

Topic Area: Memory, Learning & Knowledge Structures